The in-flight catering industry is a major contributor to food wastage. This wastage is a direct result of the deliberate overproduction of in-flight meals to protect against meal shortages and dissatisfied passengers. With the global strive towards sustainability and the resulting impact of wastage on a company's corporate image, in-flight catering companies need a solution that strives to achieve zero waste and a 100% passenger satisfaction level. This dissertation evaluates the value of combining product substitution and demand uncertainty within an inventory decision-making model as a potential solution opportunity for the wastage dilemma faced by the in-flight catering industry. The decision-making model's purpose is to assist in-flight caterers to make improved decisions regarding the quantity of each meal type to produce for the specific flight under consideration. The model developed is defined as a stochastic multi-objective mixed-integer programming model with fixed recourse and two-way, stock-out based, partial consumer-driven (static) product substitution. The model relies on the output of a forecasting model, that consists of a time-inhomogeneous Markov Chain and a multiple regression model, to forecast the probability distribution of a flight's aggregate meal demand. Due to the lack of available data from public sources, synthetic data is generated to evaluate the model developed. The model is compared against three alternative models that lack either demand uncertainty, product substitution or both to validate the value of including these elements in the decision-making model. The comparison results indicate that the inclusion of the passenger load uncertainty improves the model's average reliability to achieve a 92% minimum passenger satisfaction level with at least 9.2%. Furthermore, it is shown that the stochastic passenger load model produces an average of 2.2 fewer surplus meals per flight instance at the expense of a 3.3% lower reliability when including the substitution behaviour of passengers. This substitution model's superior waste minimisation is attributed to the model's inherent risk-pooling capabilities, and further analysis shows that the value of product substitution increases when the model becomes more constrained. It is, therefore, concluded that the value of product substitution depends on the in-flight caterer's bias towards maximising either reliability or performance. / Dissertation (MEng (Industrial Engineering))--University of Pretoria, 2021. / Council for Scientific and Industrial Research (CSIR) / Industrial and Systems Engineering / MEng / Unrestricted

Tato práce popisuje a aplikuje parametrické a neparametrické modely spolehlivosti na cenzorovaná data. Ukazuje implementaci spolehlivosti v metodologii Six Sigma. Metody jsou využity pro přežití/spolehlivost reálných technických dat.

Phylogenetic tree inference is a very common method for visualising evolutionary relationships among species. This work focuses on explanation of mathematical theory behind molecular phylogenetics as well as design of a modified algorithm for phylogenetic tree inference based on intra-group analysis of nucleotide and amino acid sequences. Furthermore, it describes the object design and implementation of the proposed methods in Python language, as well as its integration into powerful bioinformatic portal. The proposed modified algorithmic solutions give better results comparing to standard methods, especially on the field of clustering of predefined groups. Finally, future work as well as an application of proposed methods to other fields of information technology are discussed.

In this thesis, we present a detailed analysis of hybrid point-to-multipoint free space optical (FSO)/radio frequency (RF) wireless system. Hybrid FSO/RF sys- tems have emerged as a promising solution for high data rate wireless transmission. FSO technology can be used effectively in multiuser scenarios to support Point-to- Multi-Point (P2MP) networks. In this P2MP network, FSO links are used for data transmission from a central location to multiple users. When more than one FSO link fail, the central node uses a common backup RF link to transmit a frame to a remote node using an equal priority protocol. An equal priority protocol means that the remote nodes have the same priorities in being assigned the RF link. We assume two traffic classes, a high-priority and low-priority classes. The base station reserves two transmit buffers of each user for the downlink transmission. Considering the downlink traffic from the base station to a tagged remote node, we study several performance metrics. We develop a cross-layer Markov chain model to study the throughput from central node to a remote node as well as the performance of the resulting system. / Graduate

Tato diplomová práce se zabývá odhadováním matice intenzit Markovova pro- cesu se spojitým časem na základě diskrétně pozorovaných dat. Začátek práce je věnován jednoduššímu odhadu ze spojité trajektorie pomocí metody maximální věrohodnosti. Dále je zde popsán odhad z diskrétní trajektorie přes výpočet ma- tice pravděpodobností přechodu. Následně je velmi podrobně rozebrán EM al- goritmus, který předchozí odhad zpřesňuje. Na závěr teoretické části je uvedena metoda odhadu zvaná Monte Carlo Markov Chain. Všechny postupy jsou zároveň implementovány v počítačovém softwaru a prezentace jejich výsledk· je obsahem druhé části práce. V té jsou porovnané odhady pro denní, týdenní a měsíční po- zorování a také pro pětiletou a desetiletou pozorovanou trajektorii. K výsledk·m jsou připojeny odhady rozptyl· a intervaly spolehlivosti. 1

This thesis deals with multi-stage stochastic programming in the context of random process representation. Basic structure for random process is a scenario tree. The thesis introduces general and stage-independent scenario tree and their properties. Scenario trees can be also combined with Markov chains which describe the state of the system and determine which scenario tree should be used. Another structure which enables reduce the complexity of the problem is a scenario lattice. Scenario generation is performed using moment method. Scenario trees are used for representation of random returns as the input to the investment problem.

Colorectal carcinoma (CRC) ranks high in mortality and morbidity in most developed countries. Following theses focus on specific aspects of colorectal carcinoma pathogenesis including the issue of screening. The goal of the first study was assessment of expression of epithelial markers of colorectal carcinogenesis p53, COX-2, bcl-2. The study included patients with active ulcerative colitis (UCA), ulcerative colitis in remission (UCR), primary sclerosing cholangitis with ulcerative colitis (PSC-UC) (PSC), patients after liver transplantation for PSC (OLT) and a control group (N). We found significantly increased expression of tumour suppressor gene p53 in non-dysplastic mucosae in PSC-UC compared with UCA, UCR, OLT, and N, which may indicate higher neoplastic potential of PSC. Statistically significant correlation was found between PSC incidence and p53 expression. Surprisingly, OLT showed no p53 expression in non-dysplastic mucosa compared with PSC-UC. This indicates that PSC may contribute to increased expression of p53 and p53-induced colorectal carcinogenesis. Furthermore, a correlation between expression of p53 and COX-2 together with the increased expression of bcl-2 in UCA compared to N can support the role of inflammation in colorectal carcinogenesis. The goal of the second study was...

This thesis deals with multi-stage stochastic programming in the context of random process representation. Basic structure for random process is a scenario tree. The thesis introduces general and stage-independent scenario tree and their properties. Scenario trees combined with Markov chains are also introduced. Markov chains states determine if there is a crisis period or not. Information about historical number of crises helps us to construct a scenario lattice. Scenario generation is performed using moment method. Scenario trees are used as an input to the investment problem.

Parameter inference for stochastic systems is considered as one of the fundamental classical problems in the domain of computational systems biology. The problem becomes challenging and often analytically intractable with the large number of uncertain parameters. In this scenario, Markov Chain Monte Carlo (MCMC) algorithms have been proved to be highly effective. For a stochastic system, the most accurate description of the kinetics is given by the Chemical Master Equation (CME). Unfortunately, analytical solution of CME is often intractable even for considerably small amount of chemically reacting species due to its super exponential state space complexity. As a solution, Stochastic Simulation Algorithm (SSA) using Monte Carlo approach was introduced to simulate the chemical process defined by the CME. SSA is an exact stochastic method to simulate CME but it also suffers from high time complexity due to simulation of every reaction. Therefore computation of likelihood function (based on exact CME) in MCMC becomes expensive which alternately makes the rejection step expensive. In this generic work, we introduce different approximations of CME as a pre-conditioning step to the full MCMC to make rejection cheaper. The goal is to avoid expensive computation of exact CME as far as possible. We show that, with effective pre-conditioning scheme, one can save a considerable amount of exact CME computations maintaining similar convergence characteristics. Additionally, we investigate three different sampling schemes (dense sampling, longer sampling and i.i.d sampling) under which convergence for MCMC using exact CME for parameter estimation can be analyzed. We find that under i.i.d sampling scheme, better convergence can be achieved than that of dense sampling of the same process or sampling the same process for longer time. We verify our theoretical findings for two different processes: linear birth-death and dimerization.Apart from providing a framework for parameter inference using CME, this work also provides us the reasons behind avoiding CME (in general) as a parameter estimation technique for so long years after its formulation

I takt med digitaliseringen blir datadrivet beslutsfattande det nya normala i många branscher. Konkurrensfördelarna är allmänt kända eftersom det hjälper företag att utvecklas. Denna fallstudie syftar till att belysa de möjligheter som datadriven optimering bidrar med för skidorter när det kommer till att förbättra tjänster och anpassa skidanläggningar för framtiden. Huvudfokuset är att studera rörelsemönster hos skidåkare med hjälp av processutvinningsverktyg och andra metoder för visualisering. Detta har lett till följande forskningsfrågor: Vilken information går att utvinna ur data från liftsystem? Hur skulle denna typ av information kunna skapa värde i en organisation? Tidigare studier inom detta forskningsområde visar på stora möjligheter med användning av datautvinning och uppmanar till fortsatt forskning. Studien bidrar till forskningen genom att studera specifika åldersgrupper vilket tidigare inte genomförts. Studien visar att det finns skillnader i rörelsemönster hos olika åldersgrupper av skidåkare, vilket i sin tur visar på potentiella optimeringsområden hos skidanläggningarna. Utöver att belysa potentiella förbättringsområden med hjälp av datadrivna beslut visar studien även på en markant förändring hos typen av skidåkare som besöker svenska skidorter 2021, vilket troligtvis berodde på att Alperna höll stängt under skidsäsongen. I framtiden kan studien spela en viktig roll för forskning gällande hur Covid-19 påverkade svenska skidorter. / Given the digitalization, data-driven decision making is becoming the new normal in many industries. The competitive advantages are widely known as it helps companies to evolve. This case study aims to highlight the possibilities data-driven optimization provides when it comes to improving services and adapting to the future for ski resorts. Our focus is skier movement patterns which we generated by analyzing ski lift transportation data with a process mining tool and other methods for visualizations. Hence, our research questions: What information can be extracted from lift usage data? In what way can this information create value in an organization? Previous studies done in the field demonstrate many possibilities with data mining and urges for continued research. The research provided by this study is a contribution to the field through the research done on specific age-groups as this has not previously been done. This study introduces findings based on differences in the movement patterns based on skier age groups which lead to possible areas of optimization. In addition to highlighting possible ways to improve decision making using data, this study shows a significant shift in the type of skier visiting the Swedish ski-resorts 2021, possibly due to The Alps being closed this season. In the future, this study could play an essential role in studying how Covid-19 impacted Swedish ski-resorts.